The present invention is directed to a micro-organism, which is able to degrade hydantoins to enantiomerically enriched amino acids. Especially, this micro-organism is equipped with cloned genes coding for the necessary enzymes.
Racemic 5-monosubstituted hydantoins can be chemically synthesized according to Bucherer-Berg method using aldehydes, ammonium bicarbonate and sodium cyanide as reactants. They are important precursors for the enzymatic production of D- and L-amino acids. With the increasing demand for optically pure amino acids a lot of effort has been made towards the isolation of microorganisms capable for stereospecific hydrolysis of the hydantoins and characterization of the enzymes (Syldatk and Pietzsch, xe2x80x9cHydrolysis and formation of hydantoinsxe2x80x9d (1995), VCH Verlag, Weinhein, pp. 403-434; Ogawa et al., J. Mol. Catal. B: Enzym 2 (1997), 163-176; Syldatk et al., Appl. Microbiol. Biotechnol. 51 (1999), 293-309). The asymmetric bio-conversion to either L- or D-amino acids consists of three steps:
(i) chemical and/or enzymatic racemization of 5-substituted hydantoins
(ii) ring opening hydrolysis achieved by a hydantoinase and
(iii) hydrolysis of the N-carbamoyl amino acid produced by hydantoinase to the amino acid by carbamoylase.
Arthrobacter aurescens DSM 3747 is one of the few isolated microorganisms capable of converting 5-monosubstituted hydantoins to L-amino acids. The disadvantage of using A. aurescens cells as biocatalyst is the low enzyme activity. Especially the L-N-carbamoylase is the bottleneck for most substrates leading to an increase of the intermediate L-N-carbamoyl amino acid in the cell, which is not further converted to the corresponding amino acid. By combining the purified enzymes bottlenecks could be avoided but due to the low amounts of enzymes in the cells and loss of activity during the many necessary purification steps this process is not cost-effective.
All three genes encoding for the racemase hyuA (SEQ ID NO: 11), the L-specific hydantoinase hyuH (SEQ ID NO: 9) and the stereoselective L-N-carbamoylase (SEQ ID NO: 7) have been cloned in E. coli separately and expressed to high levels (about 10% of the total cell protein) (DE 19913741; J. Biotechnol., to be published). For in vitro catalysis the enzymes from the three recombinant strains can be produced and purified more cost-effective then from the Arthrobacter aurescens strain. Regarding the different enzyme activities towards the various substrates the enzymes can be combined in enzyme reactors at ratios optimized for each reaction.
It is an object of this invention to provide a further possibility of how a racemase, a hydantoinase and a D- or L-specific carbamoylase can act together in a process for the production of enantiomerically enriched amino acids from 5-monosubstituted hydantoins. Especially, this possibility should be suitable to be implemented in processes on technical scale, that is to say it has to be most cost-effective.
This is done by using a whole cell catalyst according to claim 1. Further preferred catalysts are subjects to claims depending from claim 1. Claims 6 to 9 are directed to a process for the production of the whole cell catalyst of the invention. Claims 10 and 11 protect a process for the production of enantiomerically enriched amino acids using the catalyst according to the invention.
Using whole cell catalysts comprising cloned genes encoding for a hydantoinase, for a hydantoin racemase and a D- or L-specific carbamoylase for the conversion of 5-monosubstituted hydantoins to L- or D-amino acids results in a fast and complete conversion of racemic mixtures of hydantoins to the corresponding L- or D-amino acids on industrial scale. This significantly reduces the production costs due to a reduction of fermentation and purification costs because all enzymes are produced in one strain.
Advantageously, a bacteria is used as cell, because of high reproduction rates and easy growing conditions to be applied. There are several bacteria known to the skilled worker which can be utilized in this respect. Preferably a Escheria coli can be used as cell and expression system in this regard (Yanisch-Perron et al. Gene (1985), 33, 103-109).
It is another positive embodiment of this invention that in principle all genes encoding for the hydantoinase, racemase and carbamoylase known to the artisan can be taken to be expressed in the whole cell catalyst. Preferably all genes can be taken from DSM 3747 (SEQ ID NO: 7, 9, 11).
The enzymes to be incorporated in the genetic code of the whole cell catalyst naturally possess different turnover rates. It is a drawback if the rates of co-working enzymes are not in line and intermediates accumulate during the production inside the cell. The overexpression of the hydantoinase gene in E. coli leads to the formation of inclusion bodies (Wiese et al., in preparation), which is unfavourable for a well balanced coexpression of all the three enzymes. Therefore, various attempts to xe2x80x9cfine tunexe2x80x9d the expression of these genes have been made. This can be done advantageously by overexpressing the hydantoinase genes in question according to their turnover rates. According to the DSM 3747-System the hydantoinase gene is overexpresses from plasmids with reduced copy numbers.
A further embodiment of the instant invention is directed to a process for the production of the whole cell catalyst according to the invention. In principle all plasmids known to the skilled worker can serve to carry the gene into the expression system. Preferably, plasmids derived from pSC101, pACYC184 or pBR322 are used to produce the catalyst. Most preferably plasmids pBW31 and pBW32, pBW34 and pBW35, pBW34 and pBW53, pBW32 or pBW34 are used in this respect. For the skilled worker plasmids and methods to produce plasmids can be deduced from Studier et al., Methods Enzymol. 1990, 185, 61-69 or brochures of Novagen, Promega, New England Biolabs, Clontech or Gibco BRL. More applicable plasmids, vectors can be found in: DNA cloning: a practical approach. Volume I-III, edited by D. M. Glover, IRL Press Ltd., Oxford, Washington D.C., 1985, 1987; Denhardt, D. T. and Colasanti, J.: A surey of vectors for regulating expression of cloned DNA in E. coli. In: Rodriguez, R. L. and Denhardt, D. T (eds), Vectors, Butterworth, Stoneham, Mass., 1987, pp179-204; Gene expression technology. In: Goeddel, D. V. (eds), Methods in Enzymology, Volume 185, Academic Press, Inc., San Diego, 1990; Sambrook, J., Fritsch, E. F. and Maniatis, T. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. They are incorporated by reference herewith.
Over-expression can be accomplished be means known to the skilled artisan, e.g. using constitutive or inducible expression systems as reviewed by Makrides (Makrides, 1996, Microbiol. Rev. 60, no. 3, 512-538)
Preferably, for expression of the enzymes a rhamnose inducible E. coli promoter cassette is used.
In addition, primers useful for the amplification of the gene of the invention in a PCR are protected similarly. Primers which are feasible are for example, primers S988 (SEQ ID NO: 6), S2480 (SEQ ID NO: 1), S2248 (SEQ ID NO: 2), S2249 (SEQ ID NO: 3), S2517 (SEQ ID NO: 4) or S2518 (SEQ ID NO: 5). Furthermore, all other primers which could serve to carry out this invention and which are known to the artisan are deemed to be useful in this sense. The finding of a suitable primer is done by comparison of known DNA-sequences or translation of amino acid sequences into the codon of the organism in question (e.g. for Streptomyces: Wright et al., Gene 1992, 113, 55-65). Similarities in amino acid sequences of proteins of so called superfamilies are useful in this regard, too (Firestine et al., Chemistry and Biology 1996, 3, 779-783). Additional information can be found in Oligonucleotide synthesis: a practical approach, edited by M. J. Gait, IRL Press Ltd, Oxford Washington D.C., 1984; PCR Protocols: A guide to methods and applications, edited by M. A. Innis, D. H. Gelfound, J. J. Sninsky and T. J. White. Academic Press, Inc., San Diego, 1990. Those strategies are incorporated by reference herewith.
Another aspect of the invention is a process for the production of enantiomerically enriched amino acids, which utilizes a whole cell catalyst according to the invention. Furthermore, a process is preferred that is performed in an enzyme-membrane-reactor (DE 19910691.6).
To adopt the turnover rate of all enzymes expressed in the whole cell catalyst to each other there are different methods to achieve this.
a) The genes are expressed with different promoters. The gene with the lowest activity is combined with the strongest promoter and vice versa. A disadvantage would be that for each gene a different inductor is necessary to induce the expression of all genes.
b) The genes are expressed with one promoter on a polycistronic messenger. The ratios of synthesis of the enzymes is influenced by changing or by mutation of the translation initiation region of each gene (the ribosomal binding site) which determines the efficiency of protein synthesis. This principle is realized in operons of microorganisms. The disadvantage is that the efficiency of a translation initiation region can not be predicted which means that for each gene many changes in the translation initiation region have to be made and tested (Grifantini et al., 1998, Microbiology, 144, 947-954).
c) The enzyme activity of each enzyme can be changed by mutation using error prone PCR (Fromant et al., 1995, Anal. Biochem. 224, 347-353) and DNA shuffling (Stemmer, 1994, Nature 370, 389-391). Again, this is very time consuming and costly.
d) Instead of mutagenizing genes to optimize their function in a reaction cascade, genes from different origins which encode enzymes with appropriate properties could be combined to an operon. This needs a large database describing such enzymes.
e) All genes are expressed from the same promoter but from replicons with different copy numbers. This can be the chromosome (single copy) or plasmids with low, moderate and high copy numbers. By constructing various compatible plasmids with different copy numbers and antibiotic markers carrying each the same cassette with the promoter and a polylinker sequence, genes of interest can be integrated into the plasmids in one step and the plasmids combined within one strain. This method allows a fast construction and testing of many combinations and with just one inducer in one fermentation all genes are expressed at different levels according to the plasmid copy number.